Games crashing on new computer

Just got a new computer recently and i was testing it out on counterstrike source

im trying to run it in 1920x1200 with everything maxed out as high as it will go.
16x aa, 16x a.filtering

my frames pretty much stay a steady 60 but im crashing within 10-20 minutes usually into the game with high settings


most of the time my game will just freeze for about 1 second and go straight to desktop saying hl2 just had unexpected error or something
it almost seems like its getting overloaded from trying to load too many sounds or something

im starting to get blue screens of death after playing games for awhile as well mentioning :

"Driver IRQL_NOT_LESS_OR_EQUAL"

or

"MEMORY MANAGEMENT"

im running vist 64 premium

intel core 2 duo cpu E8400 @3.00 Ghz 6mb L2

2 gigs of ram (i think like pc 6400)

2x 8800 gt nvidia graphics cards with SLI

Asus P5N32-e Sli motherboard
onboard audio
635 watt power supply
any help would be great


UPDATE: i tried running cs at 1920x1200 with no AA or and just trifiltering with everything basically low and not fancy- getting over 100 frames but game still crashed after about 10-12 minutes

this time with a new blue screen of death message:

something like- "the video memory manager has encountered and error"

Looks like a memory problem, or overheating, check your temps, wouldn't hurt to run a memory diag either. But since this is happening after a specific interval while gaming, I'm betting on temps being the issue.

Same thought I had, temps being the main issue and then memory being another option.

alright thanks ill see if i can check it out when i get home

Last night when i was trying to figure out whats up with it i saw some temperatures

the front of the case said 33 degrees celcius
i glanced at the bios for a second and i think it was 20 degrees for the processor and 40 degrees for the mother board

is there a way to check how hot the ram or graphics cards are getting?

alright thanks ill see if i can check it out when i get home

Last night when i was trying to figure out whats up with it i saw some temperatures

the front of the case said 33 degrees celcius
i glanced at the bios for a second and i think it was 20 degrees for the processor and 40 degrees for the mother board

is there a way to check how hot the ram or graphics cards are getting?
I am with Hash and K- on memory or heat prob..

33 and 40C are not high temps; for some reason I bet those aren't accurate... What kinda of tower you got?? What kinda of cooling solution/fans(intake & exhaust) in Tower?? Do you have a nice tight clear cable wiring to allow good air flow?? What is your heatsink on your CPU??

Please give us full hardware specs;i.e., type and brand of Tower, number of HDs, Rom Drives, flat cables or rounded, yadda yadda, thanks..

You should be able to put your hand at the back of your Tower and feel the temperature of the Vid Card exhaust fans... If very warm to hot, then you got a heat problem...

heres my specs from the cyberpower receipt:

RAM-
INTEL CORE 2 DUO 8400 3GHZ RETAIL

Harddrives-
500 GB SATA II 3.0 GB 16MB 7200RPM

RAM-
2x 1 gig DDR2 PC6400 MEMORY (i remember only selecting the cheapest kind something like cosair value select- could be a factor?)

Case-
X-TELSTAR JR G TYPE / GREEN/ WINDOW -it has a display in the front

motherboard-
C2D ASUS P5N32-E SLI QUAD 680I 775

power supply-
SIGMA SHARK 635WATT POWER SUPPLY

sound-
AC97 ONBOARD SOUND

graphics cards-
2x DOTAC GEFORCE 8800GT

Cabling and fans-
ASETEK 120MM WATERCOOLER
24" IDE ROUND CABLE WITH 3 HEADS

Im actually really happy with how neat it looks in the case
i paid like 20 bucks for a "professional" guy to make it look nice
its all tied all with ties and stuff
i think in total (including the watercooling fan) i have 3 fans

Misc-
GIGABYTE 802.11G 54MBP WIRELESS PCI
12-IN-ONE INTERNAL CARD READER
BLACK LG 20X DVDRW
BLACK SONY 16X DVD
WINDOWS VISTA HOME PREMIUM 64-BIT

sorry i couldnt get the names of everything

Your PSU is low end.
It shows a lot of variance on the primary 12V rail while under load without SLI.
With 2x8800GT, I imagine that it eventually gives too much of a drop while running both cards, and that throws your error.
It is also being reviewed for a lot of excess heat while under idle loads.

Get something of a quality brand next time. PSUs you don't skimp on, ever.

Power is an issue with the system, running sli along with high end parts is going to put the power supply through it's paces. That could result in errors from the video cards or the memory.

Either way, I suggest taking out a video card and running games again. If it's a power issue it should help, if it's heat related, it should also allow for more airflow between the cards. The top card is usually cooler since the bottom card has the top blowing it's hot air directly on its back, usually way to close on some boards.

alright thanks for the input im going to take out a card and see how everything does

yeah i shoulda looked into PSUs before i picked that one out
i wasn't trying to skimp on it, i was just straight up ignorant.

thanks everyone for your help and time trying to fix my computer problem
it is greatly appreciated

if you want a quality but still pretty cheap PSU, go with Rosewill (check newegg)

I've had 2 and they were both quality 500+ watt PSU's. Corsair makes fantastic PSUs as well, but they're pretty expensive.

NEVER buy an Antec. Their rails are unstable and are pretty much the Nike's of PSUs (overpriced but they have the name)

My bet is PSU

If you have a question about a certian company:
Official XS Tiered PSU Manufacturer Brand Listing Phase III - XtremeSystems Forums (http://www.xtremesystems.org/forums/showthread.php?t=108088)

List of PSUs. It's not mine, some may disagree, but for the most part the guy is spot on.

I personally have a Thermaltake Toughpower 700W. If I were you, with the SLI 8800's I'd shoot for a better "name" brand, and a higher wattage power supply then what you currently have.

yeah its most likely the PSU

i took one of the cards out and everything works great (so far several hours no problems)
thanks everyone for all the help
ill look into getting a nicer PSU

ThermalTake 700w is a great idea. Been using thermaltake in every system, no issues.

if you want a quality but still pretty cheap PSU, go with Rosewill (check newegg)

I've had 2 and they were both quality 500+ watt PSU's. Corsair makes fantastic PSUs as well, but they're pretty expensive.

NEVER buy an Antec. Their rails are unstable and are pretty much the Nike's of PSUs (overpriced but they have the name)
lol wut?

The TruPower Trio series has some of the most stable rail levels on three rails while under full load.
The NeoPower series is much the same.
In fact, almost all PSUs made by them have about the same track record, if you are within their wattage limits. Some you can extend quite a bit more beyond the rated and still have solid line voltage.

There is a reason why in most PSU shootouts, the Antecs are all in the top quadrant.

Antec is pricey, for certain. But they're pricey because they make PSUs that last and last, with quality.

My last PSU was an Antec 450W Earthwatts. I had it running in a system that was consistently overdrawn by 50W, and not uncommon to be overdrawn by 100 or more. It lasted almost two years at that kind of stress.

Well, I hope your prob is just a PSU issue, nothing more... ;)

Guess what comes to my mind is that removing one Vid Card 'may' have cut down on the heat problem in that area of the mobo and allowed the tolerances to stay within operable parameters... Can you let us know what happens with a good higher wattage PSU and both Vid Cards installed?? thanks..

BTW, my Antec PSUs have always worked well too although I presently use the the Thermaltake 750W...

1) everybody's suggestions are a good start, but they're not going about things properly. you can't just take random stabs at things like this. you have to be very methodical. first thing's first. if you are overclocking anything, knock all the speeds back down to bone stock.

2) you should be looking for programs to actively LOG and MONITOR your temperatures as well as voltages. when your system crashes, and you've rebooted, you should be looking at those logs immediately to see what the last temperatures and voltages were. i'm not sure for the geforce what's available, but for ATI video cards, ati tray tools allowed me to log most of that information. YOU REALLY SHOULD DO THIS FIRST.

3) the BIOS and motherboard software don't monitor temperatures accurately for your cpu. or more correctly, they don't tell you the important temperatures. they monitor some other part of the cpu for the temperature. what you should really be looking at is the CORE temperature. for that, you want this program:
Core Temp (http://www.alcpu.com/CoreTemp/)

this is how temperature readings for a core 2 duo processor come about:
http://www.tomshardware.com/forum/221745-29-core-temperature-guide

4) Rosewill is cheap but decent. Don't confuse that with quality. Antec is overpriced, but still good. But before I would even recommend those 2, I would first go with FSP (Fortron). They are BY FAR the best power supply manufacturers, and many other quality power supplies are actually just relabeled FSP power supplies.

I downloaded that core temp program you suggested invar

the cores are hovering around 46-47 degrees celcius
i guess that means its ok?

i dont have anything that tracks temp and voltages at the moment

I've experienced this problem on a couple platforms. Initially I had this problem about 4 years ago. I did all the trouble shooting I could, temps were all fine, voltages fine, parts were all made by well known manufacturers, memory tested, swapped out, memtested and everything, and it still occurred. At that time my computer was starting to show its age, so I decided it was time to upgrade anyway. Ended up getting a new motherboard, CPU, and memory. Kept the same video card. The problem stopped.

4 years later it started happening to me again -- only much worse. Also, all of the old hardware was eventually replaced aside from the previously mentioned motherboard, CPU, and memory. Uptime was averaging around 1 minute. Again, I tested everything out, swapped out parts for other ones, etc. I uninstalled my motherboard and video card drivers, which help improve the uptime significantly (40 minutes to a couple hours, from 1 minute), but the problem still occurred.

Now, I'm burnt out on computers and can't be bothered to track down 4 year old replacement parts, so I decided it was a good time to upgrade.

Be sure to post if you can pinpoint what was causing your IRQL errors. I've searched the Internet endlessly and haven't found any 100% sure-fire causes yet. How has your uptime been lately?

I downloaded that core temp program you suggested invar

the cores are hovering around 46-47 degrees celcius
i guess that means its ok?

i dont have anything that tracks temp and voltages at the momentthat's 46-47 while idle? because that's hot. here, get this:
techPowerUp! :: Download Intel Thermal Analysis Tool (http://www.techpowerup.com/downloads/392/mirrors.php)

It's the Intel Thermal Analysis Tool. Once you install and run it, you'll see something that lists your CPUs... That's really your cores. You'll see a line that says "Workload level 100%" and a Start button next to it. When you hit that Start button, it will heat your processor up REALLY quickly, so you will want core temp running already. You want to hit the start button for both of your cores.

Ignore the temperature readings in TAT because they're off a tiny bit.

If your temperatures in core temp read anything even close to 75C while you're doing this, hit the stop buttons IMMEDIATELY.



I've experienced this problem on a couple platforms. Initially I had this problem about 4 years ago. I did all the trouble shooting I could, temps were all fine, voltages fine, parts were all made by well known manufacturers, memory tested, swapped out, memtested and everything, and it still occurred. At that time my computer was starting to show its age, so I decided it was time to upgrade anyway. Ended up getting a new motherboard, CPU, and memory. Kept the same video card. The problem stopped.

4 years later it started happening to me again -- only much worse. Also, all of the old hardware was eventually replaced aside from the previously mentioned motherboard, CPU, and memory. Uptime was averaging around 1 minute. Again, I tested everything out, swapped out parts for other ones, etc. I uninstalled my motherboard and video card drivers, which help improve the uptime significantly (40 minutes to a couple hours, from 1 minute), but the problem still occurred.

Now, I'm burnt out on computers and can't be bothered to track down 4 year old replacement parts, so I decided it was a good time to upgrade.

Be sure to post if you can pinpoint what was causing your IRQL errors. I've searched the Internet endlessly and haven't found any 100% sure-fire causes yet. How has your uptime been lately?

That's because the IRQ_less_than_or_equalbla hlblahllalbahl errors are really generic error messages. What causes that error on one PC probably isn't what's causing it on another.

OK. So in your experience, what were the most frequent causes? Simple hardware failures and deficiencies?